メソアンサンブルデータを活用した線状降水帯発生予測手法の開発

Abstract

In recent years, heavy rain disasters caused by line-shaped rainband have become a major problem. Early evacuation is important in order to protect human life, but it is difficult to accurately predict the occurrence of line-shaped rainband, which is necessary for evacuation decisions. In this study, we developed a method for extracting line-shaped rainband from the predicted rainfall of meso-ensemble model output by elliptical approximation. As a result, we were able to achieve 85.1% missing rate (predicted to occur, but not actually occur) and 5.4% occurrence prediction rate. Compared with the existing method, missing rate was 10.5% lower and occurrence prediction rate was 17.5% lower. It is considered that the utilization of this method for evacuation decisions contributes to the reduction of heavy rain disasters.